1
Supporting Text S1. RT-PCR analysis of Muc16 homozygous mutant testes
2
3
All of the adult tissues screened by RT-PCR in the Muc16 homozygous mutants were
4
negative for 3 different regions of the Muc16 locus except for the testes (Figure 2). To
5
understand the transcripts generated by the Muc16-targeted allele from homozygous
6
mutant testes, we performed RT-PCR, using various sets of primers (Supporting Table
7
S1, Supporting Figure S1A). A robust band of the correct size was detected using
8
primers for exons 1 and 3 in wild-type and homozygous mutant testes (Supporting
9
Figure S1B). This suggests that transcription from the Muc16 targeted locus to generate
10
mRNA containing exons 1 to 3 appears to be normal. Using an exon 3 primer present in
11
both the wild-type and null alleles and an exon 3 primer located within the exon 3 deleted
12
region of the targeted allele, we detected a positive signal in wild-type but not
13
homozygous mutant testes (Supporting Figure S1B). These results support our Southern
14
analysis that indeed the majority of exon 3 has been deleted by our targeting strategy
15
(Figure 1D). Thus, the targeted allele cannot generate exon 3-containing transcripts for
16
the region that was deleted. To determine if Muc16-lacZ chimeric transcripts were
17
generated, we used exon 3 and lacZ primers. Muc16-lacZ chimeric transcripts were
18
detected from the targeted Muc16 allele but the signal was very weak (Supporting
19
Figure S1B). In addition, using lacZ primers, lacZ transcripts downstream of the Muc16-
20
lacZ fusion were undetectable (Supporting Figure S1B). This suggests that Muc16-lacZ
21
chimeric transcripts may be very unstable, leading to insufficient production of -
22
galactosidase for detection by immunofluorescence and X-gal staining (data not shown).
23
We also performed RT-PCR using primers for exons 2 and 4 in case exon 3 which is very
1
large was skipped by alternative splicing, however, no signal of the predicted size was
2
detected (data not shown). Even if exon 3 was skipped it would lead to a frameshift and
3
no MUC16 protein should be generated. We also performed RT-PCR using primers
4
located in exons 4 and 5, and exons 5 and 6. Both sets of primers amplified the predicted
5
sized bands in both wild-type and homozygous mutant testes (Supporting Figure S1C).
6
This suggests that exons 4-6 are being transcribed in the mutant. The neo gene has its
7
own promoter (Pgk) for expression in mouse ES cells. It is possible that there may be
8
readthrough of the pA signal from the neo cassette. Therefore, we also used primers for
9
neo and exon 4. No signal was detected in homozygous mutant testes (Supporting
10
Figure S1B). This suggests that the neo pA signal is functional. Finally, we repeated the
11
RT-PCR using primers for exons 6 and 10 located downstream of the exon 3 targeted
12
modification. A robust signal was detected in wild-type testes and a detectable though
13
weaker signal in homozygous mutant testes (Supporting Figure S1B). These results are
14
similar to our initial survey of expression (Figure 2). Taken together, these results
15
suggest that the targeted allele does not express the full complement of transcripts
16
generated by the wild-type allele. In addition, no MUC16 protein was detected by
17
immunofluorescence using a polyclonal antibody (Figure 3). With respect to the testis,
18
the targeted modification is clearly a loss-of-function allele. Formal demonstration that
19
the targeted mutation is a null allele in the testis is hampered because Muc16 is a very
20
large gene with many exons [22].
21
22